CN115858876B - Follow-up content intelligent pushing method and system based on disease knowledge graph - Google Patents
Follow-up content intelligent pushing method and system based on disease knowledge graph Download PDFInfo
- Publication number
- CN115858876B CN115858876B CN202310143168.7A CN202310143168A CN115858876B CN 115858876 B CN115858876 B CN 115858876B CN 202310143168 A CN202310143168 A CN 202310143168A CN 115858876 B CN115858876 B CN 115858876B
- Authority
- CN
- China
- Prior art keywords
- resources
- follow
- reading
- recommended
- patient
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Electrically Operated Instructional Devices (AREA)
Abstract
The application relates to the technical field of disease advertising knowledge pushing, and discloses a follow-up content intelligent pushing method and system based on a disease knowledge graph, wherein the method comprises the following steps: constructing a ventilating and teaching knowledge graph, a similar text model, an optimal recommended reading mode model and an intelligent recommended model which are provided with pictures, texts, audios and videos, integrating current reading resources, similar resources and intelligent recommended resources into follow-up ventilating and teaching contents, and pushing the contents to mobile equipment of a patient; the method is characterized in that the reading behavior of a patient on follow-up education content is obtained and input into an optimal recommended reading mode model and an intelligent recommended model for optimal training, the optimal reading mode is recommended based on patient information and real data, the advertising reading rate is improved, image-text resources should be recommended in the time of the office work, audio and video resources should be recommended by a patient with reading disorder, video or image-text is suitable for diet advertising, and the like, and additional follow-up education content with high similarity and high relevance can be provided.
Description
Technical Field
The application relates to the technical field of disease advertising knowledge pushing, in particular to an intelligent follow-up content pushing method and system based on a disease knowledge graph.
Background
In recent years, more and more medical enterprises and hospitals begin to pay attention to patient education, and the patient education is not only in the meaning of the surface, but also can improve the compliance of patients through very specialized knowledge, accurate medical information and medical knowledge which cannot be searched out on the internet. In the actual treatment process, the problem of poor patient compliance is caused by the reasons of increased age, lower cultural level, inaudible diet of patients, negative attitude to adverse reaction, unclear knowledge of diseases and the like, and the treatment effect and the quality of life of the diseases are seriously affected. Therefore, patient education has important roles and social significance in disease treatment.
The patent application number 202210902893.3 discloses a multi-source information fusion-based ventilating and teaching pushing method and system, and based on the characteristics of original electronic medical records, the patient personality scale in the ventilating and teaching applet and eye movement tracking information when reading ventilating and teaching knowledge patterns are considered, so that the patient portrait is more three-dimensional, the compliance model is more accurate, but the following defects still exist:
1. more friendly reading resources cannot be recommended according to the habit of a user and the scene when things happen, image-text resources should be recommended in the working hours, audio and video resources should be recommended by patients with dyskinesia, and the diet ventilating teaching is suitable for videos or images and texts;
2. the core of the medical device is that when the compliance of the patient is insufficient, the patient is reminded to improve the compliance in a declarative manner, but the medical knowledge of the patient on the condition of the patient cannot be learned in science popularization and related content.
Disclosure of Invention
The purpose of the application is to overcome the defects of the prior art and provide a follow-up content intelligent pushing method and system based on a disease knowledge graph.
In a first aspect, a method for intelligently pushing follow-up content based on a disease knowledge graph is provided, including:
constructing an announced and taught knowledge graph with pictures, texts, audios and videos, wherein the announced and taught knowledge graph represents a semantic network of an entity and an entity relationship in a triple structure;
calculating the similarity of graphic-text type learning knowledge in the learning knowledge graph, and setting the similarity between audio and video types of the same learning resource to be the same as the similarity of the graphic-text type;
constructing an optimal recommendation reading mode model and an intelligent recommendation model;
generating a pushing task through a follow-up system ventilating and teaching plan and a patient outpatient/discharge record, wherein the matched ventilating and teaching knowledge graph;
based on patient information data, current reality data and pushing tasks, a recommended reading mode is obtained through the optimal recommended reading mode model;
inputting a pushing task and a recommended reading mode into the propaganda and education knowledge graph to acquire current reading resources;
matching similar resources of the current reading resources from the ventilating and teaching knowledge graph according to the similarity;
inputting the historical reading behavior and the current reading resource of the patient into an intelligent recommendation model to obtain intelligent recommendation resources;
integrating the current reading resources, similar resources and intelligent recommended resources into follow-up education content, and pushing the follow-up education content to mobile equipment of a patient;
and acquiring the reading behavior of the patient on the follow-up education content, and inputting the reading behavior into an optimal recommended reading mode model and an intelligent recommended model for optimal training.
Furthermore, the optimal recommended reading mode model input comprises patient information, real data and announced and taught resource information, and the current reading mode which is most suitable for the patient is calculated after the causal judgment of the Bayesian network.
Further, the intelligent recommendation model input comprises historical reading behaviors of the patient, and relevant education knowledge is calculated by combining current reading resources through a Bayesian network.
Further, the announced plan is a follow-up plan in which the follow-up system provides health announcements for the away patient.
Further, the pushing task and the recommended reading mode are input into the education knowledge graph, including: combining the announced and taught knowledge patterns, taking a pushing task as a subject of the announced and taught knowledge patterns, recommending a reading mode as a predicate of the announced and taught knowledge patterns, then taking out the announced and taught actual resources through the announced and taught knowledge patterns, taking the taken out announced and taught actual resources as an object of the announced and taught knowledge patterns, and combining the subject, predicate and object of the announced and taught knowledge patterns into the current reading resources.
Further, matching similar resources of the current reading resources from the announced knowledge graph according to the similarity, including: and taking out the first n pieces of declaration knowledge meeting the similarity threshold from the declaration knowledge graph as similar resources, wherein n is a positive integer.
Further, inputting the historical reading behavior and the current reading resource of the patient into the intelligent recommendation model comprises the following steps: and inputting the historical reading behaviors of the patient and the current reading resources into an intelligent recommendation model, predicting the classification of the learning knowledge which the patient wants to acquire, and then inputting the classification of the learning knowledge and the current reading resources into the learning knowledge map to obtain the intelligent recommendation resources.
In a second aspect, a follow-up content intelligent pushing system based on a disease knowledge graph is provided, including:
the follow-up system is used for making a follow-up plan after the patient leaves the hospital and pushing follow-up education contents after diagnosis;
the reading mode prediction module is used for predicting a recommended reading mode according to the patient information data, the current reality data and the pushing task;
the content prediction module is used for predicting intelligent recommended resources according to the historical reading behaviors and the current reading resources of the patient, wherein the intelligent recommended resources comprise associated recommended resources and similar recommended resources;
the follow-up education content recommendation module is used for integrating current reading resources, similar resources and intelligent recommended resources into follow-up education content;
and the optimization training module is used for acquiring the reading behaviors of the patient on the follow-up education content, and inputting the reading behaviors into the optimal recommended reading mode model and the intelligent recommended model for optimization training.
In a third aspect, a computer readable storage medium is provided, the computer readable medium storing program code for execution by a device, the program code comprising steps for performing the method as in any one of the implementations of the first aspect.
In a fourth aspect, there is provided an electronic device comprising a processor, a memory and a program or instruction stored on the memory and executable on the processor, which when executed by the processor implements a method as in any of the implementations of the first aspect.
The application has the following beneficial effects:
1. the propaganda and education knowledge graph in the method has association and association capability of the propaganda and education knowledge graph and can be continuously perfected;
2. based on patient information and real data, the optimal reading mode is recommended, the announced and taught reading rate is improved, image-text resources should be recommended in the working hours, audio and video resources should be recommended by patients with dyskinesia, and the diet announced and taught is suitable for videos or images and texts;
3. besides the follow-up task's propaganda and education knowledge, provide the additional follow-up propaganda and education content of high similarity and high relevance, both can regard as the supplement of follow-up propaganda and education content in the present period, also can promote the understanding of patient to the propaganda and education knowledge to can be according to the reading action of patient to follow-up propaganda and education content, optimize training to best recommendation reading mode model and intelligent recommendation model, thereby can make the recommended content satisfy patient's demand more.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application, illustrate and explain the application and are not to be construed as limiting the application.
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a follow-up content intelligent pushing method based on a disease knowledge graph according to an embodiment of the present application;
fig. 2 is a schematic diagram of the follow-up contents in the intelligent follow-up content pushing method based on the disease knowledge graph according to the first embodiment of the present application;
fig. 3 is a graph of a learning knowledge graph in a follow-up content intelligent pushing method based on a disease knowledge graph according to the first embodiment of the present application;
fig. 4 is a model diagram of an optimal recommended reading mode in the follow-up content intelligent pushing method based on a disease knowledge graph according to the first embodiment of the present application;
fig. 5 is an intelligent recommendation model diagram in a follow-up content intelligent pushing method based on a disease knowledge graph according to an embodiment of the present application;
fig. 6 is a block diagram of a follow-up content intelligent pushing system based on a disease knowledge graph according to a second embodiment of the present application.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
The first embodiment of the application relates to a follow-up content intelligent pushing method based on a disease knowledge graph, which comprises the following steps: constructing an announced and taught knowledge graph with pictures, texts, audios and videos, wherein the announced and taught knowledge graph represents a semantic network of an entity and an entity relationship in a triple structure; calculating the similarity of graphic-text type learning knowledge in the learning knowledge graph, and setting the similarity between audio and video types of the same learning resource to be the same as the similarity of the graphic-text type; constructing an optimal recommendation reading mode model and an intelligent recommendation model; generating a pushing task through a follow-up system ventilating and teaching plan and a patient outpatient/discharge record, wherein the matched ventilating and teaching knowledge graph; based on patient information data, current reality data and pushing tasks, a recommended reading mode is obtained through the optimal recommended reading mode model; inputting a pushing task and a recommended reading mode into the propaganda and education knowledge graph to acquire current reading resources; matching similar resources of the current reading resources from the ventilating and teaching knowledge graph according to the similarity; inputting the historical reading behavior and the current reading resource of the patient into an intelligent recommendation model to obtain intelligent recommendation resources; integrating the current reading resources, similar resources and intelligent recommended resources into follow-up education content, and pushing the follow-up education content to mobile equipment of a patient; the reading behavior of the patient on the follow-up education content is obtained and is input into an optimal recommended reading mode model and an intelligent recommended model for optimization training, and the education knowledge graph in the method has association and association capability of the education knowledge graph and can be continuously perfected; based on patient information and real data, the optimal reading mode is recommended, the announced and taught reading rate is improved, image-text resources should be recommended in the working hours, audio and video resources should be recommended by patients with dyskinesia, and the diet announced and taught is suitable for videos or images and texts; besides the follow-up task's propaganda and education knowledge, provide the additional follow-up propaganda and education content of high similarity and high relevance, both can regard as the supplement of follow-up propaganda and education content in the present period, also can promote the understanding of patient to the propaganda and education knowledge to can be according to the reading action of patient to follow-up propaganda and education content, optimize training to best recommendation reading mode model and intelligent recommendation model, thereby can make the recommended content satisfy patient's demand more.
Specifically, fig. 1 shows a flowchart of a follow-up content intelligent pushing method based on a disease knowledge graph in the first embodiment of the application, including:
s101, constructing an announced and taught knowledge graph with pictures, texts, audio and video, wherein the announced and taught knowledge graph represents a semantic network of an entity and an entity relationship in a triple structure;
specifically, according to the announced knowledge base of the in-hospital follow-up system, corresponding image-text resources, audio resources and video resources are respectively manufactured, and further according to the announced resources, disease catalogues, medicine catalogues and the like, an announced knowledge map is constructed, the announced knowledge map represents a semantic network of entities and entity relations by a triple structure (subject-predicate-object), the announced knowledge map takes diseases, symptoms, diet, medicines, nursing, actions, image-text announcements, audio announcements, video announcements and the like as main bodies (subjects, objects), and the association relation among the main bodies is taken as predicates (such as medicine treatment), as shown in fig. 3.
Wherein the declarative classification comprises diseases, symptoms, diets, medicines, nursing, actions and the like, diseases such as liver cirrhosis, diets such as sodium restriction salts, and the constructed triads are as follows: (liver cirrhosis (subject) and diet contraindication (predicate) and limiting sodium salt (object) and (liver cirrhosis (subject) and graphic text (predicate) and graphic text for liver cirrhosis; htm (object);
wherein, the propaganda and education content contains picture and text propaganda and education, audio propaganda and education of video, and the triplet of construction is: (limitation sodium salt (subject) -video (predicate) liver cirrhosis limitation sodium salt mp4 (object);
follow-up education and education inquiry after diagnosis: the follow-up system generates a personal follow-up task (personTask) based on a follow-up plan (plan), the reading mode prediction model calculates an optimal reading mode (readType), and the announced knowledge graph receives the personal follow-up task (personTask) and the optimal reading mode (readType), for example: the personal follow-up task (personTask) =liver cirrhosis disease announcements, the best reading mode (readType) =image-text, deduces that the best resource is liver cirrhosis image-text description, htm, and further renders the best UI display on the mobile end of the patient according to the best reading mode (readType) =image-text;
similar text model content recommendation query: taking feedback (return) of the doc2vec model as a main body query of the advertising knowledge graph, establishing an association relation between the advertising knowledge graph and feedback (return) of a similar text model, and triggering optimal recommended reading mode model prediction and subsequent steps when the behavior of a patient is changed (detailed contents of the recommendation are checked);
recommendation query of related advertising model content: according to the teaching classification recommendation calculated by the Bayesian causal network model, the diseases (subjects) and the teaching classification (objects) are deduced in combination, and new teaching resources are further deduced, for example: classification of diet and disease can be deduced according to causal network by liver cirrhosis to avoid emotional agitation and fatigue mp4, and classification of liver cirrhosis diet and liver cirrhosis description mp4 can be deduced according to the propagative knowledge graph.
S102, calculating the similarity of graphic-text type learning knowledge in the learning knowledge graph, and setting the similarity between audio and video types of the same learning resource to be the same as the similarity of the graphic-text type;
specifically, the graphics context resources in step S101 are taken, doc2vec is used to calculate vectors { vect1, vect2, & ltv. vectn }, then similarity between graphics context advertisements (vect) is calculated, cosij=similarity (vect, vectj), and further, similarity between audio and video formats of the same advertising resources is set as cosij;
wherein, the doc2vec can be used to calculate the similar text and construct a similar text model for predicting the similar text, the construction of the similar text model comprises the following steps:
s201, word segmentation is carried out on the to-be-trained ventilating and teaching text;
s202, removing the deactivated word segmentation based on the removal stop word list;
s203, inputting the filtered segmented document and the comparison document into the Doc2Vec, and iteratively updating the Doc2Vec to obtain a final document vector output by the Doc2 Vec;
s204, setting a word vector context distance window=5, sentence vector dimension vectorsize=200, minimum word frequency mincount=1, and training round number=100;
s205, doc2vec model training structure, i.e., text vector of document (vect 1, vect2,..vectn);
s206, comparing text vectors of the 2 documents by using a cosine similarity calculation function, and obtaining the similarity between the texts;
cos＝similarity(vect1,vect2)
wherein cos is a text similarity matching result, similarity is a cosine function, and vect1 and vect2 are file vectors of two comparison files.
S103, constructing an optimal recommendation reading mode model and an intelligent recommendation model;
specifically, the best recommended reading mode model is used for providing accessible and easier-to-read resources for patients, so as to further improve the reading rate of the announcements, and inputs information (such as gender, usual places, reading disorders and the like) of the patients, real data (such as workdays, daytime or evening) and information (such as the length of the announced resources, the classification of the announced resources and the like) of the announced resources, and the best suitable current reading mode of the patients is calculated after causal judgment of a Bayesian network, as shown in fig. 4;
the best recommended reading mode model is based on patient information and real data, and is used for recommending the best reading mode, improving the announced and taught reading rate, the Bayesian network comprises patient information, announced and taught knowledge information, the present diagnosis information, real information and the like, and the patient information collects the tendency of the patient to announced and taught resources and comprises: whether reading disorder, sex, usual place, occupation, usual reading mode, etc., the announced knowledge information is used for analyzing the acceptability of different forms of the announced knowledge, including: the image-text length, the audio length, the video length, the education classification and the like, the present treatment information is the treatment information corresponding to the follow-up task, and the method comprises the following steps: whether to be hospitalized, whether to be outpatient, whether to be operated, and the like, the real information is various real time and scenes, and simulates different reading selections of people in various environments, including the current time, whether to work day, and the like;
chain law of probability of joint probability function:
Pr(G,S,R)＝Pr(G|S,R)Pr(S|R)Pr(R)
wherein G, S and R are nodes with association relationship, such as usual places, professions and reading modes.
In addition, each classification of the announced and taught resources comprises causal relationships, such as medicines and diets, and the related announced and taught knowledge is further deduced by using a Bayesian network and combining the historic reading content of the patient, as shown in fig. 5;
wherein, intelligent recommendation model interpretation: the announced resources each category contains causal relationships such as drugs and diets. The related declaration content is presumed by using a Bayesian causal network, wherein the Bayesian network nodes comprise gender, age, symptoms, treatment time, current reading content classification, usual place, diagnosis, medicine preparation, historical reading classification, knowledge classification and the like, and the nodes are logically associated according to causal relations, such as the symptoms of diseases are correspondingly influenced by different sexes and ages;
chain law of probability of joint probability function:
Pr(G,S,R)＝Pr(G|S,R)Pr(S|R)Pr(R)
wherein G, S and R are nodes with association, such as gender, symptoms and diagnosis.
S104, generating a pushing task through a follow-up system' S ventilating and teaching plan and a patient clinic/discharge record, and matching ventilating and teaching knowledge maps;
specifically, the announced plan is a follow-up plan for which the follow-up system provides a healthy announced for an out-of-hospital patient (out-of-hospital/in-hospital), exemplary: liver cirrhosis discharge education and liver transplantation how to prevent infection and the like, and the corresponding education knowledge map education tasks of the patient are generated according to the education program and combined with outpatient/discharge records of the patient, for example: liver cirrhosis patients, discharge instruction on the day of discharge, and regular diet instruction after discharge.
S105, acquiring a recommended reading mode through the optimal recommended reading mode model based on patient information data, current reality data and pushing tasks;
specifically, the manner of preferentially providing the best-fit current patient reading knowledge announced in step S104 is implemented, and exemplary: the reading disorder personnel is preferentially provided with audio and video resources or the staff on the workday is more willing to accept graphic and text announcements and the like, and each factor (sex, whether the workday is, the announce type is and the like) influencing the selection of the reading mode is input into the best recommended reading mode model in the step S103, and the result is the current best reading mode of the present announce knowledge for the patient.
S106, inputting the pushing task and the recommended reading mode into the education knowledge graph to acquire current reading resources;
specifically, in combination with the announced knowledge graph, the result of step S104 is the subject of the announced knowledge graph, the result of step S105 is the predicate of the announced knowledge graph, and then the announced actual resource, i.e., the object of the announced knowledge graph, is taken out through the announced knowledge graph.
S107, matching similar resources of the current reading resources from the ventilating and teaching knowledge graph according to the similarity;
illustratively, the first n pieces of the announced knowledge with the similarity threshold value > 0.8 in step S106 are taken from the announced knowledge graph as similar knowledge, where n is a positive integer, for example: n may be 3, 5, 6, etc.
S108, inputting the historical reading behaviors and the current reading resources of the patient into the intelligent recommendation model to acquire intelligent recommendation resources;
specifically, the various types of diseases are related to each other, for example: diet, medication, diet, exercise and the like, and the patient has knowledge requirements on relevant announced knowledge of each life cycle of the disease, reading records of the patient and current reading resources in the step S106 are input into the intelligent recommendation model in the step S103, the announced knowledge classification which the patient wants to acquire is predicted, and the announced knowledge classification and the current reading resources are further input into an announced knowledge map, so that intelligent recommended resources (namely intelligent recommended announced knowledge content) are obtained.
S109, integrating the current reading resources, the similar resources and the intelligent recommended resources into follow-up education content, and pushing the follow-up education content to the mobile equipment of the patient;
the resources in steps S106, S107 and S108 are integrated and pushed to the mobile terminal device of the patient, where the pushing mode may be a mode of public number, short message, multimedia message or mailbox, and the pushed graphic form is shown in fig. 2.
S110, acquiring the reading behaviors of the patient on the follow-up education content, and inputting the reading behaviors into an optimal recommended reading mode model and an intelligent recommended model for optimal training.
Specifically, the reading behavior of the patient on the announced and taught content is obtained and input into the best recommended reading mode model and the intelligent recommended model in step S103, so as to further optimize the best recommended reading mode model and the intelligent recommended model.
Example two
Referring to fig. 6, a follow-up content intelligent pushing system based on a disease knowledge graph according to a second embodiment of the present application includes:
the follow-up system is used for making a follow-up plan after the patient leaves the hospital and pushing follow-up education contents after diagnosis;
the reading mode prediction module is used for predicting a recommended reading mode according to the patient information data, the current reality data and the pushing task;
the content prediction module is used for predicting intelligent recommended resources according to the historical reading behaviors and the current reading resources of the patient, wherein the intelligent recommended resources comprise associated recommended resources and similar recommended resources;
the follow-up education content recommendation module is used for integrating current reading resources, similar resources and intelligent recommended resources into follow-up education content;
and the optimization training module is used for acquiring the reading behaviors of the patient on the follow-up education content, and inputting the reading behaviors into the optimal recommended reading mode model and the intelligent recommended model for optimization training.
Example III
A computer readable storage medium according to a third embodiment of the present application stores program code for execution by a device, the program code including steps for performing the method in any one of the implementations of the first embodiment of the present application;
wherein the computer readable storage medium may be a Read Only Memory (ROM), a static storage device, a dynamic storage device, or a random access memory (random access memory, RAM); the computer readable storage medium may store program code which, when executed by a processor, is adapted to carry out the steps of a method as in any one of the implementations of the first embodiment of the present application.
Example IV
An electronic device according to a fourth embodiment of the present application includes a processor, a memory, and a program or an instruction stored in the memory and executable on the processor, where the program or the instruction implements a method according to any one of the implementations of the first embodiment of the present application when executed by the processor;
the processor may be a general-purpose central processing unit (central processing unit, CPU), microprocessor, application specific integrated circuit (application specific integrated circuit, ASIC), graphics processor (graphics processing unit, GPU) or one or more integrated circuits for executing relevant programs to implement the methods according to any of the implementations of the first embodiment of the present application.
The processor may also be an integrated circuit electronic device with signal processing capabilities. In implementation, each step of the method in any implementation of the first embodiment of the present application may be implemented by an integrated logic circuit of hardware in a processor or an instruction in software form.
The processor may also be a general purpose processor, a digital signal processor, an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (field programmable gatearray, FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component. The disclosed methods, steps, and logic blocks in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present application may be embodied directly in a hardware decoding processor or in a combination of hardware and software modules in the decoding processor. The software modules may be located in a random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, etc. as well known in the art. The storage medium is located in a memory, and the processor reads information in the memory, and in combination with hardware thereof, performs functions required to be performed by units included in the data processing apparatus according to the embodiment of the present application, or performs a method in any implementation manner of the first embodiment of the present application.
The above is only a preferred embodiment of the present application; the scope of protection of the present application is not limited in this respect. Any person skilled in the art, within the technical scope of the present disclosure, shall cover the protection scope of the present application by making equivalent substitutions or alterations to the technical solution and the improved concepts thereof.
Claims (9)
1. The follow-up content intelligent pushing method based on the disease knowledge graph is characterized by comprising the following steps of:
constructing an announced and taught knowledge graph with pictures, texts, audios and videos, wherein the announced and taught knowledge graph represents a semantic network of an entity and an entity relationship in a triple structure;
calculating the similarity of graphic-text type learning knowledge in the learning knowledge graph, and setting the similarity between audio and video types of the same learning resource to be the same as the similarity of the graphic-text type;
constructing an optimal recommendation reading mode model and an intelligent recommendation model;
generating a pushing task through a follow-up system ventilating and teaching plan and a patient outpatient/discharge record, wherein the matched ventilating and teaching knowledge graph;
based on patient information data, current reality data and push tasks, a recommended reading mode is obtained through the optimal recommended reading mode model, wherein the current reality data comprises workdays, daytime or nights;
inputting a pushing task and a recommended reading mode into the propaganda and education knowledge graph to acquire current reading resources, wherein the pushing task and the recommended reading mode are input into the propaganda and education knowledge graph, and the method comprises the following steps of: combining the announced and taught knowledge patterns, recommending a reading mode to be a predicate of the announced and taught knowledge patterns by taking a pushing task as a subject of the announced and taught knowledge patterns, then taking out the announced and taught actual resources through the announced and taught knowledge patterns, taking the taken out announced and taught actual resources as an object of the announced and taught knowledge patterns, and combining the subject, predicate and object of the announced and taught knowledge patterns into a current reading resource;
matching similar resources of the current reading resources from the ventilating and teaching knowledge graph according to the similarity;
inputting the historical reading behavior and the current reading resource of the patient into an intelligent recommendation model to obtain intelligent recommendation resources;
integrating the current reading resources, similar resources and intelligent recommended resources into follow-up education content, and pushing the follow-up education content to mobile equipment of a patient;
and acquiring the reading behavior of the patient on the follow-up education content, and inputting the reading behavior into an optimal recommended reading mode model and an intelligent recommended model for optimal training.
2. The intelligent follow-up content pushing method based on the disease knowledge graph according to claim 1, wherein the optimal recommended reading mode model input comprises patient information, real data and announced and taught resource information, and the current reading mode which is most suitable for the patient is calculated after the causal judgment of the Bayesian network.
3. The intelligent pushing method for follow-up content based on disease knowledge graph according to claim 2, wherein the intelligent recommendation model input comprises historical reading behavior of a patient, and related education knowledge is calculated by combining current reading resources through a Bayesian network.
4. The intelligent pushing method for follow-up contents based on disease knowledge graph according to claim 1, wherein the follow-up plan is a follow-up plan for providing health education for the patient away from hospital by a follow-up system.
5. The disease knowledge graph-based follow-up content intelligent pushing method according to claim 1, wherein matching similar resources of a current reading resource from the announced knowledge graph according to the similarity comprises: and taking out the first n pieces of declaration knowledge meeting the similarity threshold from the declaration knowledge graph as similar resources, wherein n is a positive integer.
6. The disease knowledge graph-based follow-up content intelligent pushing method according to claim 1, wherein inputting the historical reading behavior and the current reading resources of the patient into the intelligent recommendation model comprises: and inputting the historical reading behaviors of the patient and the current reading resources into an intelligent recommendation model, predicting the classification of the learning knowledge which the patient wants to acquire, and then inputting the classification of the learning knowledge and the current reading resources into the learning knowledge map to obtain the intelligent recommendation resources.
7. A pushing system based on the disease knowledge-graph-based follow-up content intelligent pushing method as claimed in any one of claims 1-6, comprising:
the follow-up system is used for making a follow-up plan after the patient leaves the hospital and pushing follow-up education contents after diagnosis;
the reading mode prediction module is used for predicting a recommended reading mode according to the patient information data, the current reality data and the pushing task;
the content prediction module is used for predicting intelligent recommended resources according to the historical reading behaviors and the current reading resources of the patient, wherein the intelligent recommended resources comprise associated recommended resources and similar recommended resources;
the follow-up education content recommendation module is used for integrating current reading resources, similar resources and intelligent recommended resources into follow-up education content;
and the optimization training module is used for acquiring the reading behaviors of the patient on the follow-up education content, and inputting the reading behaviors into the optimal recommended reading mode model and the intelligent recommended model for optimization training.
8. A computer readable storage medium storing program code for execution by a device, the program code comprising steps for performing the method of any one of claims 1-6.
9. An electronic device comprising a processor, a memory, and a program or instruction stored on the memory and executable on the processor, which when executed by the processor, implements the method of any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310143168.7A CN115858876B (en) | 2023-02-21 | 2023-02-21 | Follow-up content intelligent pushing method and system based on disease knowledge graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310143168.7A CN115858876B (en) | 2023-02-21 | 2023-02-21 | Follow-up content intelligent pushing method and system based on disease knowledge graph |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115858876A CN115858876A (en) | 2023-03-28 |
| CN115858876B true CN115858876B (en) | 2023-05-26 |
Family
ID=85658558
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310143168.7A Active CN115858876B (en) | 2023-02-21 | 2023-02-21 | Follow-up content intelligent pushing method and system based on disease knowledge graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115858876B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117056575B (en) * | 2023-10-12 | 2024-01-30 | 深圳市华图测控***有限公司 | Method for data acquisition based on intelligent book recommendation system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106548431A (en) * | 2016-10-14 | 2017-03-29 | 杭州伯仲信息科技有限公司 | A kind of patient propaganda and education course pushes platform and method for pushing |
| CN115376707A (en) * | 2022-07-15 | 2022-11-22 | 重庆大学附属肿瘤医院 | Personalized chemotherapy drug knowledge propaganda and education information service method, system and terminal |
| CN115408599A (en) * | 2021-05-27 | 2022-11-29 | 京东方科技集团股份有限公司 | Information recommendation method and device, electronic equipment and computer-readable storage medium |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103970753B (en) * | 2013-01-28 | 2017-06-20 | 北大方正集团有限公司 | The method for pushing and device of association knowledge |
| US20200265953A1 (en) * | 2019-02-14 | 2020-08-20 | Babylon Partners Limited | Identifying Valid Medical Data for Facilitating Accurate Medical Diagnosis |
| CN111753543B (en) * | 2020-06-24 | 2024-03-12 | 北京百度网讯科技有限公司 | Medicine recommendation method, device, electronic equipment and storage medium |
| CN114334179A (en) * | 2021-11-17 | 2022-04-12 | 宁波市第一医院 | Digital medical management method and system |
| CN114121295A (en) * | 2021-11-29 | 2022-03-01 | 复旦大学附属中山医院 | Construction method of knowledge graph driven liver cancer diagnosis and treatment scheme recommendation system |
| CN114496234B (en) * | 2022-04-18 | 2022-07-19 | 浙江大学 | Cognitive-atlas-based personalized diagnosis and treatment scheme recommendation system for general patients |
| CN114780785B (en) * | 2022-06-23 | 2022-09-13 | 新缪斯(深圳)音乐科技产业发展有限公司 | Music teaching recommendation method and system based on knowledge graph |
| CN114969557B (en) * | 2022-07-29 | 2022-11-08 | 之江实验室 | Propaganda and education pushing method and system based on multi-source information fusion |
-
2023
- 2023-02-21 CN CN202310143168.7A patent/CN115858876B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106548431A (en) * | 2016-10-14 | 2017-03-29 | 杭州伯仲信息科技有限公司 | A kind of patient propaganda and education course pushes platform and method for pushing |
| CN115408599A (en) * | 2021-05-27 | 2022-11-29 | 京东方科技集团股份有限公司 | Information recommendation method and device, electronic equipment and computer-readable storage medium |
| CN115376707A (en) * | 2022-07-15 | 2022-11-22 | 重庆大学附属肿瘤医院 | Personalized chemotherapy drug knowledge propaganda and education information service method, system and terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115858876A (en) | 2023-03-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Du et al. | ML-Net: multi-label classification of biomedical texts with deep neural networks | |
| Jain et al. | Attention is not explanation | |
| CN112992317B (en) | Medical data processing method, system, equipment and medium | |
| Lu et al. | Machine learning for synthetic data generation: a review | |
| Sager et al. | A survey of image labelling for computer vision applications | |
| US20210104316A1 (en) | Data driven analysis, modeling, and semi-supervised machine learning for qualitative and quantitative determinations | |
| US20220229855A1 (en) | Genealogical entity resolution system and method | |
| Che et al. | Multimodal federated learning: A survey | |
| CN115858876B (en) | Follow-up content intelligent pushing method and system based on disease knowledge graph | |
| Park et al. | Understanding the representation and representativeness of age in AI data sets | |
| Jayawardhana et al. | An ontology-based framework for extracting spatio-temporal influenza data using Twitter | |
| CN111986793A (en) | Diagnosis guide processing method and device based on artificial intelligence, computer equipment and medium | |
| Fatima et al. | Smart CDSS: Integration of social media and interaction engine (SMIE) in healthcare for chronic disease patients | |
| CN113345523A (en) | Microorganism-disease association prediction method and system based on graph attention network | |
| Lai et al. | Parks, green space, and happiness: a spatially specific sentiment analysis using microblogs in Shanghai, China | |
| Jing et al. | Knowledge-enhanced attentive learning for answer selection in community question answering systems | |
| Sagong et al. | The effects of smartphone use on life satisfaction in older adults: The mediating role of depressive symptoms | |
| Zhou et al. | The digital divide is aging: An intergenerational investigation of social media engagement in China | |
| Newson et al. | Robit regression in Stata | |
| CN111859117A (en) | Information recommendation method and device, electronic equipment and readable storage medium | |
| CN116578704A (en) | Text emotion classification method, device, equipment and computer readable medium | |
| Chee et al. | Semiparametric estimation for nonparametric frailty models using nonparametric maximum likelihood approach | |
| Quintero et al. | Estimation for finite mixture of simplex models: applications to biomedical data | |
| CN113656589B (en) | Object attribute determining method, device, computer equipment and storage medium | |
| Xie et al. | Exploring Media Images of Librarians in Newspapers of Mainland China and Taiwan. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |